Type 1 diabetes (T1D) is predicted by autoantibodies (Ab) directed against the 65kD isoform of glutamic acid decarboxylase (GAD65), IA-2, -a protein tyrosine phosphatase-like molecule-, and insulin. The objective is to uncover the mechanisms by which GAD65, IA-2, and insulin autoantibodies predict T1D. The overall hypothesis is that both antibody avidity and epitope specificity mature as an individual progresses towards TED reflecting antigen-driven processes. The aims are: 1. To test the hypothesis that subtype and isotype autoantibodies to GAD65, IA-2, and insulin, alone or in combination predict progression from islet autoimmunity to T1D. Specific subtype and isotype autoantibodies against GAD65, IA-2, and insulin will be analyzed as a function of progression or non-progression to T1D. 2. To test the hypothesis that progression to T1D is associated with autoantibody maturation against specific autoantigen epitopes. Using isoform and subtype specific assays with both GAD65/67 hybrid proteins and competing epitope-specific recombinant Fab, autoantibody maturation to specific epitopes will be established as a function of progression or non-progression to T1D. 3. To test the hypothesis that progression to T1D is associated with the avidity of epitope-specific autoantibodies. Titers and autoantibody affinity maturation will be examined in competition assays with recombinant Fab to unique epitopes of GAD65 and insulin. Measures of autoantibody avidity and affinity constants will be used to distinguish non-progressors from progressors to T1D. Project 2 will collaborate with a) Project 1 on non-progressors, progressors and new onset T1D patients and CD4 T cell profiling in relation to the autoantibody characteristics; b) project 3 on new onset classic T1D patients compared to type 1.5 diabetes subjects and T cell proliferation to multiple islet cell proteins and cytokine secretion patterns and c) Project 4 HLA and non-HLA genetic factors in general population subjects of T1D risk followed until islet autoimmunity and T1D. Core B will be instrumental to sequence Fab cDNA cloned from hybridoma cell lines producing monoclonal antibodies to GAD65, IA2, and insulin. We expect to find that autoantibody maturation measures will predict progression to diabetes better than the mere autoantibody-positivity.